Publication Date


Document Type


First Advisor

Polansky, Alan M.

Degree Name

M.S. (Master of Science)

Legacy Department

Department of Statistics and Actuarial Science


Systems, and their reliabilities, depend on the reliabilities of the components that theyare composed of, and in this paper we want to nd the system structure that is the most likely given observed data. Bayesian methods were utilized in order to discover the posterior means, or observed reliabilities, of both the components and the systems. Assuming the serial and parallel system structures have independent components, we calculated system reliabilities based on observed component reliabilities by using the multiplication and addi- tion probability rules. We are then able to expand upon the numerical comparison method through a maximum likelihood analysis that compares the computed system reliability to the observed system reliability. To account for any variation, we then simulate new data using the observed data that is passed through our maximum likelihood analysis in order to discover the most likely system structure. These functions are developed using R code; and our process is illustrated using an example dataset.


44 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type